Several power supply systems satisfying some of these constraints are already known by those skilled in the art.
One such power supply system is for example described in document US 2007/0018502 A, which describes an assembly for a continuous power supply system with a dual input making it possible to eliminate energy losses due to the switching function between the two power supply networks to which it is connected. In this system, the selection of the power supply network is made via a detection circuit that compares the voltages provided by the two networks and selects the power supply network based on the result of that comparison.
Another power supply system of this type is described in document U.S. Pat. No. 4,558,230, which describes a dual-input power supply system capable of switching securely between the two power supply networks.
However, these solutions are not fully satisfactory.
In fact, in the first document, the switching logic between the networks is rigid and obeys a control law depending on the voltage of the power supply networks. Furthermore, the described system does not safely avoid short-circuiting of the two power supply networks, which may have serious repercussions when that system is integrated onboard an aircraft. In fact, this device does not tolerate a simple failure, since if one of the diodes breaks in short circuit, a network may draw directly from the other.
The power supply system described in document U.S. Pat. No. 4,558,230 fully duplicates the power chain as well as most of the control circuits. This device is therefore particularly bulky and heavy.
One of the aims of the invention is therefore to obtain a secure continuous power supply system with multiple inputs delivering a regulated output voltage and having an optimized efficiency, weight, bulk and reliability.